Deedee Kato, VP of corporate marketing at Foxit, envisioned a reality in which users prompt such an agent via “smart commands” to have AI assistants complete tasks of significant magnitude. “In the future, we’re looking to make multi-step smart commands,” Kato revealed. “So, you can write a prompt that says, ‘Make this PDF court-ready,’ then it would do 10 steps.” Such multi-dimensionality, or the multi-step nature of workflows, makes them particularly tenable for KM use cases because almost anything can be incorporated into them. “It’s really common, if you’re in higher education, if it involves the admissions office, and they have a contract for a software or some vendor, that the head of that office is going to have to sign off on it if it’s above a certain value,” Allen said. “That can be encoded into a workflow.”